FR2710782A1 - Neutron tube with magnetic confinement of electrons by permanent magnets and its manufacturing process. - Google Patents

Abstract

Sealed tube (11) containing a DT gas mixture under low pressure, fitted with a target (4) an extraction-acceleration electrode (EEA) (2) and an ion source (1) with 2n Penning cells with anodes in a single cathode cavity (77) and a permanent magnetization system (SAP) (81 or 82, 86 or 87) per cell. The ions, formed in the ion source (1), through the simultaneous presence of an electric and magnetic field, are accelerated by the EEA (2) and projected onto the target (4) to cause an emission of neutrons there. . In this tube, half (n) of the SAPs (81, 86, respectively 82, 87) are magnetized at nominal value and the other half little or not magnetized. Application: neutron tubes.

Description

"Neutron tube with magnetic confinement of electrons by

  permanent magnets and its manufacturing process ".

Description:

  The present invention relates to a sealed neutron tube containing a deuterium-tritium or deuterium-deuterium gas mixture under low pressure and comprising an ion source, an extraction accelerating electrode and a target electrode, the ion source consisting of 2n cells elementary emission of Penning type, these cells having anodes disposed in a single cathode cavity provided with an extraction orifice per cell along the axis of the latter, and each cell comprising a permanent magnetization system, the ions , formed by electron bombardment on said gas mixture, in the ion source, being channeled by a magnetic confinement field created by said magnetization systems, extracted through the extraction orifices pierced in the cathode and accelerated by said extraction-accelerating electrode, and projected onto said target electrode to produce a reaction therein

  melting process resulting in neutron emission.

  The invention also relates to a method (s) of

manufacture of such a tube.

  Neutron tubes of this type are used in fast, thermal, epithermal or cold neutron examination techniques: neutronography, activation analysis, spectrometric analysis of inelastic scattering

  or radiative captures, neutron scattering, etc.

  Achieving the full effectiveness of these nuclear techniques requires, for the corresponding emission levels, durations of longer tubes, which constitutes, for the present invention, one of the

technical problems to solve.

  The melting reaction T (d, n) 4He delivering 14 MeV neutrons is usually the most used because of its large cross section for relatively low ion energies (deuterium-tritium or D-T tubes). However, whatever the reaction used, the number of neutrons obtained per unit of charge transiting in the beam, is always increasing as the energy of the ions directed towards a thick target is itself increasing and this largely beyond the energies of the ions obtained in the sealed tubes currently available and supplied by a THT not exceeding

250 kV.

  Among the main limiting factors in the life of a neutron tube, the degradation of the target is preponderant, due in part to the erosion of the target by ion bombardment, and secondly , the local bursting of the target under the effect of the accumulation of the implanted gases which do not diffuse, particularly for the tubes with strong

emission level.

  Erosion is a function of the nature of the material, and the structure of the target on the one hand, the energy of the incident ions and their density distribution profile on the

impact surface on the other hand.

  In most cases, the target is constituted by a hydrurable material (titanium, scandium, zirconium, erbium, etc ...) capable of fixing and releasing large quantities of hydrogen without significant disturbance of its mechanical strength: the quantity total fixed is a function of the temperature of the target and the hydrogen pressure in the tube. The target materials used are deposited in the form of thin layers whose thickness is limited by problems of adhesion of the layer on its support. One means of delaying the erosion of the target consists, for example, in forming the absorbent active layer of a stack of identical layers isolated from each other by a diffusion barrier. The thickness of each of the active layers is of the order of the penetration depth of the deuterium ions striking the target. This technique is known from the European patent application 0 338 619 in the name of the applicant

(PHF88 / 525).

  Another known way of protecting the target and thus increasing the life of the tube, compatible with that indicated in the previous paragraph, is to act on the ion beam so as to improve its density distribution profile on the surface. impact: at constant total ion current on the target, resulting in a constant neutron emission, this improvement will result in a distribution as uniform as possible of the current density over the entire surface

  offered by the target to ion bombardment.

  In a sealed neutron tube, the ions are generally provided by a Penning ion source which has the advantage of being robust, of having a cold cathode (hence a long duration of use), of give high discharge currents for low pressures (of the order of 10 A / torr),

  to have a high yield and to be of small dimensions.

  The aforementioned drawbacks militate in favor of a source of multicellular type, that is to say consisting of a Penning-type elementary cell structure comprising, for all the cells, a common cathode cavity as described for example in FIG. European Patent Application Publication No. 0 368 947 in the name of the Applicant (PHF 88/605). Note that the operation of the tube is performed at a constant average current value, Icm, this value being related to a maximum acceptable energy dissipation for the tube. A multicellular source operating at pressures of the order of -4 torr, retains the advantage of the tube longer life by more uniform wear of the target but other problems arise, in particular impulse operation , for which are defined a repetition frequency F associated with the period T (T = 1 / F), a duty cycle or duty ratio Tt, and a start delay T. The existence of this delay T is a parasitic effect of the tube, operating in pressure regulation, which one strives to minimize. For the highest pressure values, the peak current increases and the work rate Tt decreases in exactly the opposite proportion, to the point that a maximum value Fmax is reached when the pressure is reduced due to the existence of a Associated T value, which can be annoying because it is not possible to reach certain types of pulsed operation (high work rate and frequency) and also because the jitter effect due to random variations T is maximal. A second problem appears concomitantly for certain operating conditions, which is the difficulty of obtaining a stable loop for regulating the pressure. This is due to the fact that the pressure is a function at the power 17 of the regulation current Ir of the hydrogen reservoir incorporated in the tube, and that it is, in these conditions, difficult to generate a sufficiently small Ir deviation step, given the range to cover in a computer controlled system. The invention makes it possible to reduce this disadvantage by offering the possibility of operation at

higher pressure.

  An object of the present invention is to offer the possibility of doubling the lifetime of neutron tubes without

  impair the simplicity of realization of these tubes.

  Another goal is to achieve a multicellular neutron tube that reconciles the advantage of longer life with improved pulse operation for

  highest operating frequencies.

  These goals are achieved and the drawbacks of the prior art are mitigated thanks to the fact that the neutron tube defined in the preamble is remarkable in that among the 2n permanent magnetization systems, half (n) is magnetized at a nominal value allowing correct operation of the associated cell and the other half (n) is magnetized to a null or low value which does not allow the associated cell to function, so as to have a

  selective magnetization of ion sources.

  Preferably, each permanent magnet system extends along the axis of a cell and comprises two magnets disposed on either side of the cathode cavity, among

  which an annular magnet surrounds the extraction orifice.

  Preferably, the ion source, with a single cathode cavity, comprises a common, cylindrical anode comprising 2n anode holes, constituting the elementary cells, arranged substantially facing the extraction orifices. The basic idea of the invention consists in designing a source of Penning-type ions which, instead of having a single elementary cell, comprises 2n, which is used in half. To switch between the sources, it acts on the magnetization of the permanent magnets intended to produce the induction field necessary for the priming of the discharge, the power supply being common to all the cells of the source of ions, which is the simplicity factor mentioned above. A practical way to perform the switching is to initially equip n cells only, among the 2n, magnetization systems magnetized in a given direction, the other n other systems not being magnetized, then at the end of life of the tube, c ' that is to say when the wear of the target is such that the neutron level is lower than the specification, to magnetize all the magnetization systems in the opposite direction of the original direction, so as to activate the n cells not yet used while neutralizing the first n and thus allowing a second cycle of use of the tube, of substantially equal duration

to that of the previous cycle.

  From the point of view of the lifetime, one can then ask the question of the interest of activating successively n elementary cells operating at nominal current rather than running simultaneously the 2n cells with half current, in order to obtain the same lifetime, with the same neutron flow. In continuous operation mode, we could consider that the two solutions are equivalent if we ignore differences in density distribution profile.

  bombardment of the target, depending on the pressure.

  On the other hand, in pulsed operation - which remains the privileged field of use of a neutron generator, with respect to the isotopic source - it is necessary to take into account the delay in the priming of the ion source, which increases when the pressure decreases, which has the effect of limiting the

maximum frequency of use.

  Moreover, this use in two stages should not exclude the possible use of the tube with simultaneous operation of all the elementary sources (2n), which may be of interest in the following cases, indicated by way of illustration: To avoid the pressure zone o the operation is unstable. This possibility is to increase the range

of use of the tube.

  - To double the output level (without then benefiting from the increase in the life time) to the extent that the

  heat dissipation is compatible.

  - To choose an operating pressure which favors the stability of the neutron emission (as far as the working rate / operating frequency can be satisfied). This is because, in the pressure regulation mode, the stability of the tube current I depends on the IC AIR regulation accuracy of the IR tank current of a R

AIó AIR

  variable way, depending on the pressure I = K IR Ic IR

  with 13 K s 56 depending on the value of the pressure.

  The description that follows, next to the drawings

  annexed, all given as an example, will make it clear

  how the invention can be realized.

  FIG. 1 represents: - in A, in elevation and in section, the diagram of a sealed neutron tube according to the invention, - in B, the ion source of FIG. 1A, inverted and enlarged, - in C and in D, the magnets which constitute the magnetization system of each elementary cell,

in E, the view E-E of FIG. 1A.

  FIGS. 2A and 2B are graphs which show the comparative operation of a tube of the prior art and a

tube according to the invention.

  The diagrams in FIG. 1 show the main basic elements of a sealed neutron tube 11 enclosing a

  gaseous mixture under low pressure to ionize such as deuterium

  tritium and which comprises an ion source 1 and an extraction-acceleration electrode 2 between which there is a very high potential difference (which can exceed 150 kV) allowing the extraction and the acceleration of the ion beams 31, 32 and their projection on a target electrode 4 o is the reaction of

  fusion resulting in a 14 MeV neutron emission.

  The ion source 1 is shown enlarged and returned to FIG. It is secured to an insulator 5 which isolates it electrically from the remaining part of the tube which is

  complementary, and has a terminal 55 supply THT.

  This source of ions, preferably Penning type, consists of at least one cylindrical anode 6, and a cathode structure 7 which defines a cathode cavity 77 with which is associated a permanent magnetization system. In the figure, this system is constituted by magnets such as 81 or 82, and 86 or 87, shown enlarged in Figures 1D and 1C, axial magnetic field which confines the ionized gas 9 around each anode cylinder. An ion emission channel, also called extraction orifice 12 is made opposite each anode cylinder.

  The Penning source of Figure 1 is multi-

  cellular and, for a more optimal implementation of the invention, it preferably comprises an even number, ie 2n elementary emission cells which are all contained in the single cathode cavity 77. Each cell may comprise an anode which is clean but preferably the anode 6 is common, consisting of a metal disk 15 to 20 mm thick, pierced with as many holes as there are cells, each hole constituting a cylinder of anode in which electrons oscillate. Figure 1E, which shows the anode structure seen from above, also shows, in broken lines, the common anode 6, pierced with 6 holes 10 regularly spaced in a circle with an angular pitch of 60 '. Opposite each bore 10 is an extraction orifice represented in solid lines, 12. In FIG. 1B, above each bore 10, a permanent magnet 86, 87 respectively, extends along the axis 13 of FIG. each cell, disposed between the cathode 7 and a common yoke 14 which allows the magnetic field created axially in each cell to bounce in each pair of magnets 87, 82, respectively 86, 81, which constitutes a permanent magnetization system for each cell. In a multi-cellular Penning source of a known type, it is possible to use, as a permanent magnetization system, a single large permanent magnet arranged in the space occupied by the magnets 86 and 87 of FIGS. lB. Note also that the direction given to the magnetic field is irrelevant for proper operation of the tube since the function assigned to this field is to confer electrons oscillating in the anode holes, trajectories in helices to increase the distance traveled and thus increase the probability of collision with an atom or a molecule of hydrogen, which causes ionization. These trajectories can thus indifferently be dextrorsum or senestrorsum. For the implementation of the invention, it is necessary that each cell has its own magnetization system, preferably constituted by a pair of magnets, on either side of the anode, rather than a single one. type 86 or 87 magnet, even if the yoke 14 may, meanwhile, remain common to all magnets. The magnets 86 and 87 preferably have the shape of a cylindrical bar (FIG. 1C) approximately 30 mm long and 14 mm in diameter. The magnets 81 and 82 have a cylindrical ring shape with a thickness of 8 mm and an outside diameter of about 20 mm. These magnets are preferably based on aluminum, nickel and cobalt, such as those known under the trade name "Alnico 1500", the magnetic field & develop must be of the order of 1000 Gauss. Alnico has the advantage of supporting, without loss of induction, high temperatures (of the order of 400 C). This makes it possible to reduce the steaming time for degassing before sealing in the tube of the ion source provided with its magnets. Other types of magnets, such as Samarium magnets, are more powerful but, in the current state of technology, they lose their magnetization above a certain temperature (200 C for Samarium magnets). Samarium magnets could however be used, for the implementation of the invention, by reducing the temperature

  steaming while increasing the duration.

  Usually, for a neutron tube of the type of FIG. 1, the magnets are put in place, in the neutral state, in the yoke 14, with the other elements 7, 6, already described, and this subassembly is its turn arranged in the tube 11, inside which is introduced a partial vacuum (after degassing) and which is then sealed. It is only at this stage that the magnetization of the magnets is operated in the known art, the tube being placed on a magnetization bench. This magnetization is carried out at one time, and at saturation, by means of an induction coil arranged around the tube, traversed by a current pulse of

magnetization sufficiently long.

  According to the invention, on the contrary, the magnetization is carried out first, and a discrimination is made, a half n only magnets being previously magnetized (at saturation), while the other half is left in the neutral state , before mounting in the cylinder head 14. The mounting of the tube is then continued as indicated above, but there is no other magnetization phase before putting the tube into service. Note however that in the tube according to the invention thus developed, the magnetization of the magnets now incorporated in the tube, remains possible, in one direction or the other, by magnetization of the tube placed on an ad hoc magnetization bench. In Figure 1, magnets previously magnetized (activated) are referenced 81, 86, while the magnets such as 82, 87, are left in the neutral state. Preferably, the pattern of the activated magnetization systems (type 81 and 86) is regular, within the ion source 1, being complementary and exactly imbricated with the pattern of the non-activated magnet systems (type 82 and 87). For example, as shown in Figure 1, 6 magnetization systems are distributed along the generatrices of a cylinder with an angular pitch of 60, so that an activated magnetization system alternates with a magnetization system not activated . The magnetization systems are mounted in the tube with a given direction of magnetization, the same for all. During the commissioning of this tube according to the invention, only ion sources can operate where a fairly high magnetic field prevails, generated by the magnets 86, 81. In the other sources, the path of the electrons is insufficient to start the production of ion-electron pairs, the latter electrons can in turn produce other ionelectron pairs. Indeed, in the absence of a magnetic field, the electrons are captured by the anode without oscillating. Their course is too short to have a significant probability of achieving

ionization.

  A tube equivalent to a multi-cellular tube with n ion sources is thus obtained. After an operating time equal to T (a few hundred to a few thousand hours depending on the requested output level) the tube reaches the end of its life, by the fact that, at the location of the impact centers of ion beams on the target where the ion bombardment is the strongest, the target is worn to the point that the output level for which it was set has dropped significantly. According to the invention, the n-source tube is returned to a new initial state of operation by a carefully performed remagnetization which deactivates the n previously active ion sources and activates the other n ion sources. For this, the tube must be placed on the magnetization bench and receive a magnetization pulse, in the opposite direction to said direction given to n previously activated sources, with an intensity capable of canceling, at the first order, the magnetization of these latest. At the same time, the n sources which were up tooror in the neutral magnetic state acquire a magnetization in the opposite direction of said given direction. It will be noted that during this (final) magnetization phase, the n activated magnetization systems should not be magnetized at saturation, at the risk of obtaining a tube with the 2n magnetization systems activated according to the same (new) sense of magnetization. However, if we exclude the risk of perforation of the substrate of the target, it is possible in the latter case to nevertheless find the nominal neutron output level, at the cost of a double current, to the extent that the generator is capable of current asked and o the tube can accept double heat dissipation. Thus, the tube with n sources according to the invention can be used for a period still equal to T approximately. We thus obtain, all things being equal

  elsewhere, and in particular the internal pressure, a multi-

  cell (n sources) whose lifetime is twice that of a multi-cellular tube (n sources of the prior art). It should also be noted that the magnetization of the tube carried out at mid-life requires a fairly high precision as regards the new state of magnetization sought for all the magnets. In this context, it is favorable, in order to avoid dispersion of magnetizations, to use magnets which have similar characteristics to each other.

  as close as possible and, preferably, magnets of the same batch.

  It is indicated above that during the so-called final magnetization of the tube, the n activated magnetization systems should not be magnetized to saturation. To compensate for this slight disadvantage, which would result in n new sources of ions at a rate lower than that of the n primitively active sources, it is possible, in a variant, to oversize, in proportion to this magnetic non-saturation,

magnets of the type 82, 87.

  Furthermore, it must be considered that the amount of neutrons that can be emitted by a tube is directly related to the amount of (for example) tritium that can store the target in its (or its) layer (s) active (s). This consideration can lead to comparing the tube according to the invention to a tube of the prior art with 2n ion sources (ie the tube according to the invention but for which one would magnetize all the sources at the same time. time, before the first operation), each source would operate at half (ionic) current relative to the sources described above (ie Ic / 2n instead of Ic / n), but for a double time, it would be that is 2T. Compared with the tube of the invention, the same total target current Ic, and hence the same neutron flow, and the same lifetime, equal to 2T in the first and the second, are obtained in the first order. other case. This comparison shows that the advantages of the invention are only partially related to the neutron flow of the tube or its lifetime, and therefore little related to the total number of neutrons that the tube is able to emit during its lifetime . On the other hand, a net advantage appears, linked to the fact that the tube according to the invention operates

  at a pressure substantially double that of the multi-

  2n ion source of known art, this advantage being

  reserved for pulsed operation (or impulses).

  In pulsed operation, there is a delay at initiation, T, at each pulse. This delay, which increases when the pressure in the tube decreases, is furthermore more troublesome as it fluctuates, from one pulse to the next (phenomenon of wobbling). For a given target current Ic and a real work rate Tt (expressed in%), there is therefore an upper limit to the maximum operating frequency of the tube that can be reached. The advantage of increased pressure in pulsed operation, to reduce the duration T, appears especially for the deuterium-deuterium type (DD) type neutron tubes for which priming is more difficult because the electronic "seeds" that favor this priming is rarer than in the deuterium-tritium tube (DT) where a partial pressure of tritium exists in the chamber of the tube, and it follows that T delays and fluctuations of these delays are much greater than for

  the D-T tube (at equal pressure in both tubes).

  FIG. 2A is a double graph which indicates, in logarithmic coordinates, and for a deuterium-tritium tube, the relationships that exist between the main parameters (mentioned above) characterizing the tube in pulsed operation. In the upper graph, the 3-segment curve representative of the tube according to the invention is referenced 511, 512, 513, while the curve (2-segment) for the equivalent tube according to the known art (2n ion sources) is referenced 521, 522. By limiting itself to the first segment of each curve, 511, respectively 521, it is possible to deduce from FIG. 2A the graph of FIG. 2B on which the maximum frequency curve Fmax relating to the invention is referenced max. 515 and that relating to the known art, 525 (Fmax = - '). Each curve, set in neutron pulse duration values, ti in ps, limits upwards the operating frequency range that is physically possible to obtain with either tube. It is observed that the maximum operating frequencies Fmax, which provide the points of the curves 515 and 525, are more than doubled, for the invention, with respect to the known art. The invention applies to neutron tubes of the D-D or D-type. The invention is compatible with other improvements aimed at increasing the lifetime of the target and, consequently, of the tube. It may be mentioned in particular the tube having a target with several active layers described in the European patent O 338 619 in the name of the Applicant (PHF88 / 525),

already indicated above.

  In order to improve the life of a multi-cellular neutron tube of the type described above in the first paragraph, it may also be envisaged to use several sources of elementary ions that are each individually electrically energized individually. It would therefore be equivalent, as to the desired result, to discriminate on the power supply for the anode polarization of the cells. However, the realization of such a tube, if it can be envisaged, would lead to a much more complex realization than for the discrimination between magnetic fields, given the problems of realization of the heart of the elementary cells, with

  the electrical insulation problems attached to it.

Claims (5)

CLAIMS:   1. Sealed neutron tube containing a gaseous mixture of deuterium-tritium or deuterium-deuterium at low pressure and   having an ion source, an extraction electrode   acceleration and a target electrode, the ion source consisting of 2n elementary emission cells Penning type, these cells having anodes disposed in a single cathode cavity provided with an extraction orifice per cell along the axis of the latter, and each cell comprising a permanent magnetization system, the ions, formed by electron bombardment on said gas mixture, in the ion source, being channeled by a magnetic confinement field created by said magnetization systems extracted through the extraction orifices pierced in the cathode and accelerated by said extraction-acceleration electrode, and projected onto said target electrode to produce a fusion reaction causing a neutron emission, characterized in that among the 2n systems of permanent magnetization half (n) is magnetized at a nominal value allowing correct functioning of the associated cells and the other half (n) is magnetized to a null or low value which does not allow the associated cells to function, so as to have a selective magnetization of ion source.   2. Neutron tube according to claim 1, characterized in that each permanent magnetization system extends along the axis of a cell and comprises two magnets arranged on either side of the cathode cavity, among   which an annular magnet surrounds the extraction orifice.   A neutron tube according to claim 1 or 2, wherein said ion source, having a single cathode cavity, comprises a common, cylindrical anode having 2n anode holes, constituting the elementary cells, arranged in   substance opposite the extraction orifices.   4. Process for manufacturing a sealed neutron tube   according to one of claims 1 to 3, containing a gaseous mixture   deuterium-tritium at low pressure and comprising an ion source, an extraction-acceleration electrode and a target electrode, the ion source consisting of 2n elementary cells Penning type, all 2n cells comprising an anode common in a single cathode cavity provided with an extraction orifice per cell along the axis of the latter, and a permanent magnetization system, characterized in that the first step consists in mounting n elementary cells of the source of ions with magnetization systems already magnetized according to an initial induction that has a given meaning, the other n magnetization systems being mounted unmagnetized on the n other cells.   5. A method of manufacturing a neutron tube according to claim 4, characterized in that after reaching the end of life of the target fraction bombarded by the n magnetized cells in said given direction, a second magnetization step of predetermined value in a direction opposite to said given direction, of the whole of the ion source, concomitantly causes the decommissioning of the n cells previously   active and commissioning of the other n cells.